1 BACKGROUND

In a series of national surveys, CAMH and research technology company Delvinia led research to understand the mental health and substance use impacts of COVID-19, and to track changes as the pandemic unfolded [1]. The following interactive dashboard represents selected findings based on online surveys conducted over the following intervals:

Wave 1: May 8 to 12, 2020, with 1,005 respondents
Wave 2: May 29 to June 1, 2020, with 1,002 respondents
Wave 3: June 19 to 23, 2020 with 1,005 respondents
Wave 4: July 10 to 14, 2020 with 1,003 respondents
Wave 5: September 18 to 22, 2020, with 1,003 respondents
Wave 6: November 27 to December 1, 2020 with 1,003 respondents
Wave 7: March 19 to 23, 2021 with 1,000 respondents
Wave 8: July 9 to 13 2021, with 1,001 respondents
Wave 9: January 7 to 11, 2022, with 1,004 respondents

All survey participants were English-speaking Canadians ages 18 and older [1].

Below are some key findings for this particular study, wave 6.

Data source: CAMH Studies Mental Health over Coronavirus Pandemic

knitr::include_graphics("https://i.imgur.com/BmTwhaW.png")

Figure 1.1: The Delvinia Website

2 OBJECTIVES

In this report, I employ the different packages in R and various functions to generate firstly the initial analysis of the data, and second the subsequent steps to prepare data for analysis. I use the the dataset provided by the 6th wave of the CAMH study to generate insights regarding:

The dataset itself
The study’s research results

3 METHODOLOGY - SELECT THE DATA

In this report, I use RStudio to scrutinize the data and generate descriptive analyses using relevant packages. Some of the packages that I use and their functions are:

dplyr: A powerful and efficient data manipulation package that provides a set of tools for filtering, grouping, and summarizing data. It is particularly useful for working with large datasets and is widely used in data wrangling tasks.
ggplot2: A data visualization package that provides a flexible and powerful way to create a wide range of static and interactive plots. It is widely used for creating high-quality data visualizations and is particularly useful for exploratory data analysis.
pacman: An R package management tool, which provides a convenient way to install and manage R packages, similar to the package manager on Linux and MacOS. It is a wrapper around the basic functionality provided by the “install.packages()” function and provides a more user-friendly interface for installing, updating, and removing packages.

3.1 Import the dataset and the datamap

The data and the datamap for CAMH study wave 6th were imported.

camh <- read.csv("http://bus-sawtooth.mcmaster.ca/eH705_BLENDED_W2020/CAMH_wave6_nl.csv")
camh_dm <- read.csv("http://bus-sawtooth.mcmaster.ca/eH705_BLENDED_W2020/camh_Q.csv")

kable(camh_dm, caption = "The datamap of the survey") %>%
  kable_styling(bootstrap_options = "striped", full_width = F, position = "left", fixed_thead = T) %>%
  scroll_box(width = "800px", height = "500px")

Table 3.1: The datamap of the survey
Variable.ID	Question.Label
respid,	respid
status,	status
L,	Language
hWave,	wave tracker
Consent,	Consent
S1,	In which province or territory do you currently live?
hRegion,	In which province or territory do you currently live?
S2,	To which of the following age groups do you belong? (detailed)
hAge,	To which of the following age groups do you belong
S3,	How do you describe your gender identity? (detailed)
hGender,	How do you describe your gender identity?
Q4_1,	I have tested positive for COVID-19
Q4_2,	Someone close to me has tested positive for COVID-19
Q4_3,	I have had symptoms of COVID-19 but have not been tested
Q4_4,	Someone close to me has had symptoms of COVID-19 but has not been tested
Q4_5,	I have been tested for COVID-19 but it was negative
Q4_6,	Someone close to me has been tested for COVID-19 but it was negative
Q4_7,	I am elderly and/or have a health condition that increases the risk
Q4_8,	Someone close to me is elderly and/or has a health condition that increases the risk
Q4_9,	I have a job that exposes me to high risk
Q4_10,	Someone close to me has a job that exposes them to high risk
Q4_99,	No covid-19 risks
Q5,	How worried are you about the impact of COVID-19 on your personal financial situation?
Q6,	Physical distancing measures affected your employment situation?
Q6b,	Number of hours you are working for pay been affected
Q7,	How worried are you that you or someone close to you will get ill from COVID-19?
Q8x1,	Feeling nervous, anxious or on edge
Q8x2,	Not being able to stop or control worrying
Q8x3,	Worrying too much about different things
Q8x4,	Trouble relaxing
Q8x5,	Being so restless that it<92>s hard to sit still
Q8x6,	Becoming easily annoyed or irritable
Q8x7,	Feeling afraid as if something awful might happen
Q15,	During the past 7 days, on how many days did you drink ALCOHOL?
Q16,	On how many of the past 7 days did you drink 4 or 5 or more drinks on one occasion?
Q17,	In the past 7 days, did you drink more ALCOHOL, about same, or lessthan you did before the COVID-19?
Q18,	During the past 7 days, on how many days did you use CANNABIS?
Q19,	In the past 7 days, did you use CANNABIS ?
Q20x1,	In the past 7 days, how often have you felt depressed?
Q20x2,	In the past 7 days, how often have you felt lonely?
Q20x3,	In the past 7 days, how often have you felt hopeful about the future?
Q23CP_1,	Coping - Meditation or Mindfulness
Q23CP_2,	Coping - Physical activity
Q23CP_3,	Coping - Connected socially with friends or family online or over the phone
Q23CP_4,	Coping - Connected socially with friends or family in person
Q23CP_5,	Coping - Made time to relax
Q23CP_6,	Coping - Followed a routine
Q23CP_7,	Coping - Spent time outdoors
Q23CP_8,	Coping - Dedicated time to oneself
Q23CP_9,	Coping - Engaged in hobbies
Q23CP_10,	Coping - Sought help for mental health
Q23CP_11,	Coping - Volunteered or donated
Q23CP_12,	Coping - Engaged in spiritual practices
Q23,	Including yourself, how many people are currently living in your household?
hHouseHold,	Including yourself, how many people are currently living in your household?
Q24_1,	How many children in each of the following categories live in your household?
hChildren,	Do you have children living in your household?
Q25,	What is the highest level of education you have completed?
Q26,	What is your current marital status?
Q27,	Which of the following best describes your racial or ethnic group?
Q28,	Total household income (detailed)
hIncome,	Total household income
Q29,	Type of residence location

3.2 Trim and filter only important variables

str(camh)
view(camh)

A brief overview shows that there are 1003 observations of 66 variables in this dataframe. I will limit my work to the following 17 variables only:

hAge ( Age_group),
Gender (male, female, other),
Q4_1, Q4_3, Q4_5,
Q5 (switch coding 1->4, 2->3, 3->2, 4->1),
Q8x1 - Q8x3,
Q15,
Q17,
Q20x1, Q20x2, Q20x3,
Q25 Education
Q26 Marital status
Q27 Ethnicity/Race

camh_select <- camh %>% select(Age_group, 
                               Gender, 
                               Q4_1_me_covid_test_pos, 
                               Q4_3_me_covid_symptoms_no_test, 
                               Q4_5_me_covid_test_neg, 
                               Q5_fin_worry,
                               Q8x1_nervous_frq, 
                               Q8x2_not_stop_worry_frq, 
                               Q8x3_worry_diff_frq, 
                               Q15_alcohol, 
                               Q17_alcohol_frq, 
                               Q20x1_depressed_frq, 
                               Q20x2_lonely_frq, 
                               Q20x3_hopeful_frq, 
                               Q25_Education, 
                               Q26_marital, 
                               Q27_race)

head(camh_select) %>%
  knitr::kable(caption = "New dataframe of only 17 variables of interest") %>%
  kable_classic() %>%
  scroll_box(width = "800px", height = "300px")

Table 3.2: New dataframe of only 17 variables of interest
Age_group	Gender	Q4_1_me_covid_test_pos	Q4_3_me_covid_symptoms_no_test	Q4_5_me_covid_test_neg	Q5_fin_worry	Q8x1_nervous_frq	Q8x2_not_stop_worry_frq	Q8x3_worry_diff_frq	Q15_alcohol	Q17_alcohol_frq	Q20x1_depressed_frq	Q20x2_lonely_frq	Q20x3_hopeful_frq	Q25_Education	Q26_marital	Q27_race
1	2	1	1	1	1	2	2	2	0	3	3	4	3	5	6	8
3	1	1	1	1	1	2	2	3	2	3	2	3	2	4	1	8
1	1	1	1	1	1	2	2	2	2	5	1	2	1	3	1	8
2	2	1	1	1	2	2	2	2	96	2	3	2	1	5	1	8
3	2	1	1	1	2	1	1	4	0	5	1	1	2	2	1	8
3	2	1	1	1	2	2	2	2	99	3	2	2	3	3	3	8

I switched coding for Q5 - “How worried are you about the impact of COVID-19 on your personal financial situation?” as per instruction (switch coding 1->4, 2->3, 3->2, 4->1).

camh_new = camh_select %>% mutate(Q5_fin_worry=recode(Q5_fin_worry, '1' = 4, '2' = 3, '3' = 2, '4' = 1))

head(camh_new)%>%
  knitr::kable(caption = "New dataframe of switched codes") %>%
  kable_classic()  %>%
  scroll_box(width = "800px", height = "300px")

Table 3.3: New dataframe of switched codes
Age_group	Gender	Q4_1_me_covid_test_pos	Q4_3_me_covid_symptoms_no_test	Q4_5_me_covid_test_neg	Q5_fin_worry	Q8x1_nervous_frq	Q8x2_not_stop_worry_frq	Q8x3_worry_diff_frq	Q15_alcohol	Q17_alcohol_frq	Q20x1_depressed_frq	Q20x2_lonely_frq	Q20x3_hopeful_frq	Q25_Education	Q26_marital	Q27_race
1	2	1	1	1	4	2	2	2	0	3	3	4	3	5	6	8
3	1	1	1	1	4	2	2	3	2	3	2	3	2	4	1	8
1	1	1	1	1	4	2	2	2	2	5	1	2	1	3	1	8
2	2	1	1	1	3	2	2	2	96	2	3	2	1	5	1	8
3	2	1	1	1	3	1	1	4	0	5	1	1	2	2	1	8
3	2	1	1	1	3	2	2	2	99	3	2	2	3	3	3	8

3.3 Renaming the variables

I also rename the variable in this new dataframe to enhance clarity and comprehensiveness.

colnames(camh_new) <- c("age_group",
                        "gender",
                        "covid_positive",
                        "covid_no_test",
                        "covid_test_negative",
                        "financial_worry",
                        "nervous_frq",
                        "not_stop_worry_frq",
                        "worry_diff_frq",
                        "alcohol_drink",
                        "alcohol_frq",
                        "depressed_frq",
                        "lonely_frq",
                        "hopeful_frq",
                        "education",
                        "marital",
                        "race")
head(camh_new) %>%
  knitr::kable(caption = "New dataframe of selected variables renamed") %>%
    kable_classic()  %>%
  scroll_box(width = "800px", height = "300px")

Table 3.4: New dataframe of selected variables renamed
age_group	gender	covid_positive	covid_no_test	covid_test_negative	financial_worry	nervous_frq	not_stop_worry_frq	worry_diff_frq	alcohol_drink	alcohol_frq	depressed_frq	lonely_frq	hopeful_frq	education	marital	race
1	2	1	1	1	4	2	2	2	0	3	3	4	3	5	6	8
3	1	1	1	1	4	2	2	3	2	3	2	3	2	4	1	8
1	1	1	1	1	4	2	2	2	2	5	1	2	1	3	1	8
2	2	1	1	1	3	2	2	2	96	2	3	2	1	5	1	8
3	2	1	1	1	3	1	1	4	0	5	1	1	2	2	1	8
3	2	1	1	1	3	2	2	2	99	3	2	2	3	3	3	8

4 FINDINGS

4.1 Findings - Observations from srutinizing the data

Now the data is prepared, I carry on to examine the data, identify any missing values, or any errors in the collected data.

knitr::kable(describe(camh_new), caption = "Summary statistic of the dataframe") %>%
  kable_classic_2()

Table 4.1: Summary statistic of the dataframe
	vars	n	mean	sd	median	trimmed	mad	min	max	range	skew	kurtosis	se
age_group	1	1003	1.914257	0.8302094	2	1.892902	1.4826	1	3	2	0.1607858	-1.5327394	0.0262142
gender	2	1003	1.517448	0.5156674	2	1.511831	0.0000	1	3	2	0.1048699	-1.5731583	0.0162824
covid_positive	3	1003	1.006979	0.0832903	1	1.000000	0.0000	1	2	1	11.8268145	138.0111426	0.0026299
covid_no_test	4	1003	1.027916	0.1648151	1	1.000000	0.0000	1	2	1	5.7229353	30.7826824	0.0052041
covid_test_negative	5	1003	1.175474	0.3805614	1	1.094645	0.0000	1	2	1	1.7038136	0.9038850	0.0120164
financial_worry	6	1003	2.713858	0.9040149	3	2.767123	1.4826	1	4	3	-0.2571560	-0.7102377	0.0285447
nervous_frq	7	1003	2.000997	0.9864344	2	1.876712	1.4826	1	4	3	0.7770422	-0.3969858	0.0311471
not_stop_worry_frq	8	1003	1.818544	0.9607959	2	1.657534	1.4826	1	4	3	0.9954127	-0.0365771	0.0303376
worry_diff_frq	9	1003	1.988036	0.9643632	2	1.860523	1.4826	1	4	3	0.7442076	-0.3956563	0.0304502
alcohol_drink	10	1003	10.163509	27.0889564	1	2.099626	1.4826	0	99	99	2.9444639	6.7534390	0.8553460
alcohol_frq	11	1003	4.184447	10.5602008	3	2.938979	0.0000	1	96	95	8.5153101	71.0667061	0.3334431
depressed_frq	12	1003	1.813559	0.9520100	2	1.667497	1.4826	1	4	3	0.9330074	-0.1906952	0.0300601
lonely_frq	13	1003	1.826520	0.9908839	2	1.667497	1.4826	1	4	3	0.9235850	-0.3350125	0.0312876
hopeful_frq	14	1003	2.429711	1.0264927	2	2.412204	1.4826	1	4	3	0.0427304	-1.1416828	0.0324120
education	15	1003	5.257228	10.0462819	5	4.353674	0.0000	1	96	95	8.8104130	76.6374240	0.3172159
marital	16	1003	4.076770	11.8837134	1	2.427148	0.0000	1	96	95	7.3743845	54.1124869	0.3752336
race	17	1003	8.980060	14.7451268	8	7.069738	0.0000	1	97	96	5.5261981	29.8387681	0.4655840

camh_new_ano <- anomalies(camh_new)
knitr::kable(camh_new_ano$variables, caption = "The structure of the missing values") %>%
  kable_classic_2()

Table 4.2: The structure of the missing values
Variable	q	qZero	pZero	qDistinct	type	anomalous_percent
alcohol_drink	1003	347	34.6%	10	Integer	34.6%
covid_positive	1003	0		2	Integer
covid_no_test	1003	0		2	Integer
covid_test_negative	1003	0		2	Integer
age_group	1003	0		3	Integer
gender	1003	0		3	Integer
financial_worry	1003	0		4	Numeric
nervous_frq	1003	0		4	Integer
not_stop_worry_frq	1003	0		4	Integer
worry_diff_frq	1003	0		4	Integer
depressed_frq	1003	0		4	Integer
lonely_frq	1003	0		4	Integer
hopeful_frq	1003	0		4	Integer
alcohol_frq	1003	0		6	Integer
education	1003	0		6	Integer
marital	1003	0		7	Integer
race	1003	0		12	Integer

#create an empty list
plots <- list()

#create loops
for (i in colnames(camh_new)){
  value <- camh_new[[i]]
  variable_plot <- plot_frq(value) + labs(x = i)
  plots[[i]] <- variable_plot
}

#arrange the plots together
multiplot(plotlist = plots, cols = 5, layout = NULL)

Figure 4.1: Plot frequencies of the variables in the dataset

Observations on scrutinizing the data:

There is no N/A value in the dataframe of interest
For Q15_alcohol, “0” which can be an anomalous value in other dataset, for this partcicular case, the value is use to distinct the “non_alcohol” drinkers from “alcohol” drinkers but did not consume alcohol within the 7 days of the survey.
For some variables, existing extreme values like “95”,“96”, and “99” which skew the distribution and affecting the mean values (Such as Q15 and Q17 regarding the alcohol consumption).
For other variables in this dataframe, they are categorical, which means the value is limited and usually based on a particular finite group. However, for “Q15- During the PAST 7 DAYS, on how many days did you drink ALCOHOL?”, there are both continuous values (The number of days people drinking) and the categorical values (“I do not drink alcohol” and “Prefer not to answer”).
Therefore, for question “Q15- During the PAST 7 DAYS, on how many days did you drink ALCOHOL?”, there are three groups of respondents:
1. People who drink alcohol (Corresponding to answer from 0 - 7)
2. People who do NOT drink alcohol (Corresponding to answer 99 - “I do not drink alcohol”)
3. People who choose not to answer the question (Corresponding to answer 96 - Prefer not to answer”)
So I will break Q15 into 3 categories accordingly. And within the first category, I will maintain the numeric values for further analysis. The third category, I identify it as not applicable or n/a and thus, will proceed to remove them.
The rest of the data was fine

4.2 Findings - missing values

4.2.1 Remove n/a values

I change value “96” from Q15 into n/a and remove those n/a entries from the dataframe. This is only to demonstrate how to remove n/a from a dataset. Otherwise can use the subset() function to remove entries associated with the “96” value.

The new data frame from now will be referred to as “camh_clean”. Now there is only 999 observations

camh_clean <- camh_new

camh_clean$alcohol_drink[camh_clean$alcohol_drink == 96] <- NA
camh_clean <- na.omit(camh_clean)

kable(frq(camh_clean$alcohol_drink, out = "t"), caption = "Removing n/a values") %>%
  kable_classic() %>% kable_styling(full_width = FALSE)

Table 4.3: Removing n/a values

val	label	frq	raw.prc	valid.prc	cum.prc
0	<none>	347	34.73	34.73	34.73
1	<none>	165	16.52	16.52	51.25
2	<none>	129	12.91	12.91	64.16
3	<none>	80	8.01	8.01	72.17
4	<none>	42	4.20	4.20	76.38
5	<none>	49	4.90	4.90	81.28
6	<none>	27	2.70	2.70	83.98
7	<none>	79	7.91	7.91	91.89
99	<none>	81	8.11	8.11	100.00
NA	NA	0	0.00	NA	NA

4.2.2 Remove unnecessary data

For some questions, there are the “Preferred not to answer” option, which coded by “96”; as well as “Not sure” coded by “97”, and they occupy less than 5% of the total observations. Therefore, I will remove observations that have a value of “96” and “97” to make the data more relevant.

Now there is only 954 observations

camh_clean2 <- subset(camh_clean, !rowSums(camh_clean == 96))
camh_clean2 <- filter(camh_clean2, !rowSums(camh_clean2 == 97))
#an example: 
kable(frq(camh_clean2$race, out = "t"), caption = "Removing unnecessary data") %>%
  kable_classic() %>% kable_styling(full_width = FALSE)

Table 4.4: Removing unnecessary data

val	label	frq	raw.prc	valid.prc	cum.prc
1	<none>	129	13.52	13.52	13.52
2	<none>	42	4.40	4.40	17.92
3	<none>	24	2.52	2.52	20.44
4	<none>	12	1.26	1.26	21.70
5	<none>	13	1.36	1.36	23.06
6	<none>	9	0.94	0.94	24.00
7	<none>	16	1.68	1.68	25.68
8	<none>	679	71.17	71.17	96.86
9	<none>	11	1.15	1.15	98.01
10	<none>	19	1.99	1.99	100.00
NA	NA	0	0.00	NA	NA

4.3 Findings - Recoding

Data in the “camh_new” dataframe will be recoded, following the instructions in the new datamap, “20-3082-DATAMAP.xlsx”.

4.3.1 Recode Demographic variables

Demographic variables will be decoded directly (without creating the factor forms, as we do not need their numeric values for calculation) such as:

Age groups
Genders
Covid status (Variable 3 -5)
Education level
Marital status
Race/ Ethnicity

camh_label = camh_clean2 %>%
  mutate(age_group=recode(age_group,
                          '1' = "18-39",
                          '2' = "40-59",
                          '3' = "60+"),
         gender=recode(gender,
                       '1' = "Male",
                       '2' = "Female",
                       '3' = "Other"),
         covid_positive=recode(covid_positive,
                               '1' = "No",
                               '2' = "Yes"),
         covid_no_test=recode(covid_no_test,
                              '1' = "No",
                               '2' = "Yes"),
         covid_test_negative=recode(covid_test_negative,
                                    '1' = "No",
                                    '2' = "Yes"),
         education=recode(education,
                          '1' = "Did not graduate from high school",
                          '2' = "Completed high school",
                          '3' = "Post-high school education",
                          '4' = "College diploma/degree",
                          '5' = "University diploma/degree"),
         marital=recode(marital,
                        '1' = "Married",
                        '2' = "Living with a partner",
                        '3' = "Widowed",
                        '4' = "Divorced",
                        '5' = "Separated",
                        '6' = "Never married"),
         race=recode(race,
                     '1' = "Asian - East",
                     '2' = "Asian - South",
                     '3' = "Asian - South East",
                     '4' = "Black",
                     '5' = "Indigenous",
                     '6' = "Latin American",
                     '7' = "Middle Eastern",
                     '8' = "White",
                     '9' = "Mixed heritage",
                     '10' = "Other"))

knitr::kable(head(camh_label), caption = "Recode Demographic variables") %>%
  kable_classic()  %>%
  scroll_box(width = "800px", height = "300px")

Table 4.5: Recode Demographic variables
age_group	gender	covid_positive	covid_no_test	covid_test_negative	financial_worry	nervous_frq	not_stop_worry_frq	worry_diff_frq	alcohol_drink	alcohol_frq	depressed_frq	lonely_frq	hopeful_frq	education	marital	race
18-39	Female	No	No	No	4	2	2	2	0	3	3	4	3	University diploma/degree	Never married	White
60+	Male	No	No	No	4	2	2	3	2	3	2	3	2	College diploma/degree	Married	White
18-39	Male	No	No	No	4	2	2	2	2	5	1	2	1	Post-high school education	Married	White
60+	Female	No	No	No	3	1	1	4	0	5	1	1	2	Completed high school	Married	White
60+	Female	No	No	No	3	2	2	2	99	3	2	2	3	Post-high school education	Widowed	White
18-39	Female	No	No	Yes	4	2	2	3	2	3	2	1	2	University diploma/degree	Married	White

4.3.2 Recode Non-Demographic variables

Non-demographic variables will be decoded by creating a factor forms, namely:

Financial worry
Nervous frequency
Not stop worry frequency
Worry different things frequency
Alcohol consumption frequency
Depressed frequency
Lonely frequency
Hopeful frequency

camh_label$financial_worry_factored <- factor(camh_label$financial_worry, 
                                              levels = c(1,2,3,4), 
                                              labels = c("Not at all worried", "Not very worried", "Some what worried", "Very worried"))
camh_label$nervous_frq_factored <- factor(camh_label$nervous_frq, 
                                          levels = c(1,2,3,4), 
                                          labels = c("Not at all", "Several days", "Over half the days", "Nearly everyday"))
camh_label$not_stop_worry_factored <- factor(camh_label$not_stop_worry_frq, 
                                             levels = c(1,2,3,4), 
                                             labels = c("Not at all", "Several days", "Over half the days", "Nearly everyday"))
camh_label$worry_diff_factored <- factor(camh_label$worry_diff_frq, 
                                         levels = c(1,2,3,4), 
                                         labels = c("Not at all", "Several days", "Over half the days", "Nearly everyday"))
camh_label$alcohol_frq_factored <- factor(camh_label$alcohol_frq,
                                          levels = c(1,2,3,4,5), 
                                          labels = c("Drink much more", "Drink slightly more", "No change", "Drink slightly less", "Drink much less"))
camh_label$depressed_frq_factored <- factor(camh_label$depressed_frq,
                                            levels = c(1,2,3,4), 
                                            labels = c("Rarely or none", "Some or little", "Occasionally", "Most/All the time"))
camh_label$lonely_frq_factored <- factor(camh_label$lonely_frq,
                                            levels = c(1,2,3,4), 
                                            labels = c("Rarely or none", "Some or little", "Occasionally", "Most/All the time"))
camh_label$hopeful_frq_factored <- factor(camh_label$hopeful_frq,
                                            levels = c(1,2,3,4), 
                                            labels = c("Rarely or none", "Some or little", "Occasionally", "Most/All the time"))

knitr::kable(head(camh_label), caption = "Recode Non-Demographic variables") %>%
  kable_classic()  %>%
  scroll_box(width = "800px", height = "300px")

Table 4.6: Recode Non-Demographic variables
age_group	gender	covid_positive	covid_no_test	covid_test_negative	financial_worry	nervous_frq	not_stop_worry_frq	worry_diff_frq	alcohol_drink	alcohol_frq	depressed_frq	lonely_frq	hopeful_frq	education	marital	race	financial_worry_factored	nervous_frq_factored	not_stop_worry_factored	worry_diff_factored	alcohol_frq_factored	depressed_frq_factored	lonely_frq_factored	hopeful_frq_factored
18-39	Female	No	No	No	4	2	2	2	0	3	3	4	3	University diploma/degree	Never married	White	Very worried	Several days	Several days	Several days	No change	Occasionally	Most/All the time	Occasionally
60+	Male	No	No	No	4	2	2	3	2	3	2	3	2	College diploma/degree	Married	White	Very worried	Several days	Several days	Over half the days	No change	Some or little	Occasionally	Some or little
18-39	Male	No	No	No	4	2	2	2	2	5	1	2	1	Post-high school education	Married	White	Very worried	Several days	Several days	Several days	Drink much less	Rarely or none	Some or little	Rarely or none
60+	Female	No	No	No	3	1	1	4	0	5	1	1	2	Completed high school	Married	White	Some what worried	Not at all	Not at all	Nearly everyday	Drink much less	Rarely or none	Rarely or none	Some or little
60+	Female	No	No	No	3	2	2	2	99	3	2	2	3	Post-high school education	Widowed	White	Some what worried	Several days	Several days	Several days	No change	Some or little	Some or little	Occasionally
18-39	Female	No	No	Yes	4	2	2	3	2	3	2	1	2	University diploma/degree	Married	White	Very worried	Several days	Several days	Over half the days	No change	Some or little	Rarely or none	Some or little

4.3.3 Recode Q15 into 2 categories “Drinking” and “No drinking”

Values in question 15 will be re-coded into 2 different categories:

People who drink alcohol (Corresponding to answer from 0 - 7)
People who do NOT drink alcohol (Corresponding to answer 99 - “I do not drink alcohol”)

camh_label$alcohol_drink_cat <- if_else(camh_label$alcohol_drink %in% c(0,1,2,3,4,5,6,7), "Drinking",
                                        ifelse(camh_label$alcohol_drink %in% c(99), "Not drinking", NA))
sum(is.na(camh_label))

[1] 0

camh_label_new <- camh_label[,-(6:14)]
knitr::kable(head(camh_label_new), caption = "Clean data set ready for analysis") %>%
  kable_classic()

Table 4.7: Clean data set ready for analysis
age_group	gender	covid_positive	covid_no_test	covid_test_negative	education	marital	race	financial_worry_factored	nervous_frq_factored	not_stop_worry_factored	worry_diff_factored	alcohol_frq_factored	depressed_frq_factored	lonely_frq_factored	hopeful_frq_factored	alcohol_drink_cat
18-39	Female	No	No	No	University diploma/degree	Never married	White	Very worried	Several days	Several days	Several days	No change	Occasionally	Most/All the time	Occasionally	Drinking
60+	Male	No	No	No	College diploma/degree	Married	White	Very worried	Several days	Several days	Over half the days	No change	Some or little	Occasionally	Some or little	Drinking
18-39	Male	No	No	No	Post-high school education	Married	White	Very worried	Several days	Several days	Several days	Drink much less	Rarely or none	Some or little	Rarely or none	Drinking
60+	Female	No	No	No	Completed high school	Married	White	Some what worried	Not at all	Not at all	Nearly everyday	Drink much less	Rarely or none	Rarely or none	Some or little	Drinking
60+	Female	No	No	No	Post-high school education	Widowed	White	Some what worried	Several days	Several days	Several days	No change	Some or little	Some or little	Occasionally	Not drinking
18-39	Female	No	No	Yes	University diploma/degree	Married	White	Very worried	Several days	Several days	Over half the days	No change	Some or little	Rarely or none	Some or little	Drinking

4.4 Findings - Visualizing variables and explanation

4.4.1 Demographic profiles of the respondents to the survey wave 6th

df_1 <- camh_label_new[1:3]

demo_covid_yes <- table(subset(df_1, covid_positive == "Yes")[, c("age_group", "gender")])
demo_covid_no <- table(subset(df_1, covid_positive == "No")[, c("age_group", "gender")])


kable(demo_covid_yes, caption = "Number of respondents infected with COVID-19") %>%
  kable_classic() %>%
  kable_styling(full_width = T, position = "left") %>%
  add_header_above(bold = TRUE, align = "c", c(" " = 2, "Infected with COVID" = 1))

Table 4.8: Number of respondents infected with COVID-19
		Infected with COVID
	Female	Male
18-39	2	2
40-59	0	1
60+	0	1

kable(demo_covid_no, caption = "Number of respondents not infected with COVID-19") %>%
  kable_classic() %>%
  kable_styling(full_width = T, position = "left") %>%
  add_header_above(bold = TRUE, align = "c", c(" " = 2, "No COVID" = 2))

Table 4.9: Number of respondents not infected with COVID-19
		No COVID
	Female	Male	Other
18-39	192	169	4
40-59	145	142	1
60+	146	148	1

graph_1 <- 
  ggplot(df_1, aes(x = gender, fill = gender)) +
  geom_bar() +
  geom_text(aes(label = after_stat(count)), inherit.aes = TRUE, stat='count', vjust=-1) +
  facet_wrap(~ covid_positive, scales = "free_x") +
  labs(x = "Gender", y = "Number of respondents", fill = "Gender") +
  labs(title = "COVID-19 positive rate of survey respondents by genders") +
  theme_hc() + scale_fill_hc()
graph_2 <- 
  ggplot(df_1, aes(x = age_group, fill = age_group)) +
  geom_bar() +
  geom_text(aes(label = after_stat(count)), inherit.aes = TRUE, stat='count', vjust=-1) +
  facet_wrap(~ covid_positive, scales = "free_x") +
  labs(x = "Age groups", y = "Number of respondents", fill = "Age groups") +
  labs(title = "COVID-19 positive rate of survey respondents by age groups") +
  theme_hc() + scale_fill_hc()
multiplot(graph_1, graph_2, layout=matrix(c(1,2), nrow=1, byrow=TRUE))

Figure 4.2: Graph of COVID-19 status of survey respondents

From Table 4.8, Table 4.9, and Figure 4.2 above, we can see that the majority of people took part in this survey were not COVID-19, be it asymptomatic or have been tested. There are people with confirmed COVID-19 infection, but their number is too small to be accounted for, at 6 people over the total of 954 responses.

In terms of genders, there is an equal number of males and females taking part in this survey, with the latter group (483) is a slightly more than the former one (459). The third group of “Other” gender made up a humble statistic of 6 people.

In terms of age groups, the youngest group of respondents who are from 18 to 39 years old was significantly larger than the other two groups, with a gap as wide as 100 respondents, making up approximately 20% - 25% of the total people in each group. This may be owing to the fact that young adults are more technical savvy than their older counterparts, thus they have a greater tendency to engage in online surveys and studies similar to the CAMH research. However, it still requires further analysis to determine whether or not this difference carries any significance.

4.4.2 Investigating the difference between the drinking and not drinking groups

Investigating into the population who drink from the responses, we can see that the majority of people who participated in this survey drink alcohol.
Nevertheless, it is interesting to see that the number of males and females in each drinking and not drinking group is not significantly different.
Among people who drink, the average days on drinking in the past 7 days before taking the survey is mean(x)=1.99.
The majority of people reported that they do not drink more or less within the past 7 days, but a considerable number of people recalled to drink a slightly more during the same period.
Females are observed to have the tendency to be nervous and worried more than their male counterparts.

df_2 <- camh_label_new[,c("gender","alcohol_drink_cat")]

df_2aa <- df_2[df_2$alcohol_drink_cat == "Drinking",]
df_2bb <- df_2[df_2$alcohol_drink_cat == "Not drinking",]

df_2a <- df_2aa %>%
  group_by(gender) %>%
  tally(name = "people") %>%
  mutate(pct_a=round(people/sum(people),3)) %>%
  arrange(desc(pct_a))


df_2b <- df_2bb %>%
  group_by(gender) %>%
  tally(name = "people") %>%
  mutate(pct_b=round(people/sum(people),3)) %>%
  arrange(desc(pct_b))

tab_2a <- as.data.frame(df_2a)
tab_2b <- as.data.frame(df_2b)

kable(tab_2a, caption = "Respondents who drink") %>%
  kable_classic() %>%
  kable_styling(full_width = T, position = "left") %>%
  add_header_above(bold = TRUE, align = "c", c(" " = 1, "Drinkers" = 2))

Table 4.10: Respondents who drink
	Drinkers
gender	people	pct_a
Female	448	0.508
Male	429	0.486
Other	5	0.006

kable(tab_2b, caption = "Respondents who do not drink") %>%
  kable_classic() %>%
  kable_styling(full_width = T, position = "left") %>%
  add_header_above(bold = TRUE, align = "c", c(" " = 1, "Non drinkers" = 2))

Table 4.11: Respondents who do not drink
	Non drinkers
gender	people	pct_b
Female	37	0.514
Male	34	0.472
Other	1	0.014

df_3 <- camh_label[,c("gender","alcohol_drink","alcohol_frq_factored","nervous_frq_factored","not_stop_worry_factored","worry_diff_factored")]
df_3 <- subset(df_3, !rowSums(df_3 == 99))

df_30 <- df_3[df_3$alcohol_drink == 0,]
df_37 <- df_3[df_3$alcohol_drink == 7,]

graph3 <- df_2a %>% 
  ggplot(aes(x="", y=pct_a, fill=gender)) +
  geom_bar(stat="identity", width = 1, color="white") +
  coord_polar("y", start=0) +
  labs(title = "Respondents who drink alcohol") +
  geom_text(aes(label=paste0(pct_a*100,"%")), position = position_stack(vjust = 0.5), size=7) +
  theme_hc() + scale_fill_hc()
graph4 <- df_2b %>% 
  ggplot(aes(x="", y=pct_b, fill=gender)) +
  geom_bar(stat="identity", width = 1, color="white") +
  coord_polar("y", start=0) +
  labs(title = "Respondents who do not drink alcohol") +
  geom_text(aes(label=paste0(pct_b*100,"%")), position = position_stack(vjust = 0.5), size=7) +
  theme_hc() + scale_fill_hc()
multiplot(graph3, graph4, layout=matrix(c(1,2), nrow=1, byrow=TRUE))

Figure 4.3: Graph of drinking status of survey respondents

graph5 <- 
  ggplot(df_3, aes(x=alcohol_drink)) +
  geom_histogram(binwidth=1, fill="#009E73") +
  geom_vline(aes(xintercept=mean(alcohol_drink)), color="red", linetype="dashed") +
  geom_text(aes(label = after_stat(count)), inherit.aes = TRUE, stat='count', vjust=-1) +
  xlab("Days") +
  ylab("Frequency") +
  ggtitle("Days of drinking in the past 7 days") + 
  theme_hc() + scale_fill_hc()
graph6 <- 
  ggplot(df_3, aes(x = alcohol_frq_factored, fill = gender)) +
  geom_bar() +
  geom_text(aes(label = after_stat(count)), inherit.aes = TRUE, stat='count', vjust=-1) +
  facet_wrap(~ gender, scales = "free_x", shrink = TRUE) +
  labs(x = "Gender", y = "Number of respondents", fill = "Gender") +
  labs(title = "Frequency of alcohol consumption changes by genders") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  theme_hc() + scale_fill_hc()
show(graph5)

Figure 4.4: Graph of drinking status of survey respondents

show(graph6)

Figure 4.5: Graph of drinking status of survey respondents

4.4.3 The association of Binge drinking and stress level in respondents

A closer look into people who did not drink during the past 7 days and who drank every day regarding their stress frequency
Overall, there is no significant difference between the two groups from observing the graphs
Nonetheless, the number of people who experienced nervousness and worries for several days in the binge-drinking group accounted for the largest group.
In contrast, for people who did not drink, they appeared to do not experience any episodes of anxiety
From the graphs, people who binge_drinking might experience longer periods of stress than people who did not drink alcohol. However, more investigation is still needed to make the association of Binge drinking and stress level in respondents clearer

kable(frq(df_3$alcohol_drink, out = "t"), caption = "The structure of the number of days people drink") %>%
  kable_classic() %>%
  kable_styling(full_width = F, position = "left") %>%
  add_header_above(bold = TRUE, align = "c")

Table 4.12: The structure of the number of days people drink

val	label	frq	raw.prc	valid.prc	cum.prc
0	<none>	329	37.30	37.30	37.30
1	<none>	157	17.80	17.80	55.10
2	<none>	122	13.83	13.83	68.93
3	<none>	80	9.07	9.07	78.00
4	<none>	40	4.54	4.54	82.54
5	<none>	49	5.56	5.56	88.10
6	<none>	27	3.06	3.06	91.16
7	<none>	78	8.84	8.84	100.00
NA	NA	0	0.00	NA	NA

show(graph7)

Figure 4.6: Stress frequency among respondents

show(graph8)

Figure 4.7: Stress frequency among respondents

show(graph9)

Figure 4.8: Stress frequency among respondents

multiplot(graph7a, graph7b, graph7c, layout=matrix(c(1,2,3), nrow=1, byrow=TRUE))

Figure 4.9: Stress levels between people who did not drink and who binged drinking

multiplot(graph8a, graph8b, graph8c, layout=matrix(c(1,2,3), nrow=1, byrow=TRUE))

Figure 4.10: Stress levels between people who did not drink and who binged drinking

5 CONCLUSION

In conclusion, I would like to summarize my process for this analytic report:

I studied the data, the datamap, and filtered the 17 variables of interest
I evaluated n/a and anomalous values, weighing their impact against the value of the data set and subsequent analysis. For example, I maintained the value of 99 as “Not drinking” and used it for analysis. On the other hand, I removed values corresponding to option “Preferred not to answer” as they would not contribute to insights
I recoded the values using different methods - the factor form and the direct recoding - with respect to their nature, demographic or non-demographic values
I used different approaches - tables and graphs - to generate insights from the data set
Summary of findings:
- There are only 6 cases of COVID-19 positive out of 954 observations, making around 0.6% of the population studied
- There are equal numbers of male and female respondents
- The larger group of respondents is the youngest group of people from 18 to 39 years old
- Respondents who drink, reported to drink an average of 2 days during the past 7 days before taking the survey
- Females showed considerable tendencies to experience nervousness and worries more frequent than males. However, the relationship between this mental issues and COVID-19 need more investigation
- There is no significant difference between people who did not drink during the past 7 days and people who binged drinking the same period of time
- People who binged drinking might experience more prolonged nervous episodes but more investigation required before coming to a definitive conclusion

6 REFERENCE

[1] COVID-19 national survey dashboard. (n.d.). CAMH. https://www.camh.ca/en/health-info/mental-health-and-covid-19/covid-19-national-survey

Examining the Impact of COVID-19 on Mental Health and Substance Use among Canadians

Assignment 1 | R markdown

Author: Le Thanh Nguyen

30 January 2023